Format

Send to

Choose Destination

See 1 citation found by title matching your search:

Nucleic Acids Res. 2015 Jan;43(1):95-103. doi: 10.1093/nar/gku1288. Epub 2014 Dec 10.

Leveraging cross-link modification events in CLIP-seq for motif discovery.

Author information

1
Molecular and Computational Biology, University of Southern California, Los Angeles, CA 90089, USA.
2
Children's Cancer Research Institute and Department of Cellular and Structural Biology, University of Texas Health Science Center, San Antonio, TX 78229, USA.
3
Molecular and Computational Biology, University of Southern California, Los Angeles, CA 90089, USA uren@usc.edu.

Abstract

High-throughput protein-RNA interaction data generated by CLIP-seq has provided an unprecedented depth of access to the activities of RNA-binding proteins (RBPs), the key players in co- and post-transcriptional regulation of gene expression. Motif discovery forms part of the necessary follow-up data analysis for CLIP-seq, both to refine the exact locations of RBP binding sites, and to characterize them. The specific properties of RBP binding sites, and the CLIP-seq methods, provide additional information not usually present in the classic motif discovery problem: the binding site structure, and cross-linking induced events in reads. We show that CLIP-seq data contains clear secondary structure signals, as well as technology- and RBP-specific cross-link signals. We introduce Zagros, a motif discovery algorithm specifically designed to leverage this information and explore its impact on the quality of recovered motifs. Our results indicate that using both secondary structure and cross-link modifications can greatly improve motif discovery on CLIP-seq data. Further, the motifs we recover provide insight into the balance between sequence- and structure-specificity struck by RBP binding.

PMID:
25505146
PMCID:
PMC4288180
DOI:
10.1093/nar/gku1288
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Silverchair Information Systems Icon for PubMed Central
Loading ...
Support Center